Evaluation of an automated intracranial aneurysm detection and rupture analysis approach using cascade detection and classification networks
Ke WuDongdong GuPeihong QiXiaohuan CaoDijia WuLei ChenGuoxiang QuJiayu WangXianpan PanXuechun WangYuntian ChenLizhou ChenZhong XueJinhao Lyu
16
Citation
32
Reference
10
Related Paper
Citation Trend
Keywords:
Contextual image classification
배경: SAH의 rupture aneurysm은 다양한 부위에서 다양하게 발병한다. 특히, 두개이상의 다수의 aneurysm이 존재할 경우에는 aneurysm중 먼저 rupture되는 aneurysm을 찾아 치료하는 일은 시급히 선행되어야 할 과제이다. 목적: 따라서 본 논문에서는 다수의 aneurysm이 존재하는 경우 rupture aneurysm이 주로 발병하는 위치가 어디이며, 나아가 rupture aneurysm의 모양에 대해서 알아보고자 한다. 대상 및 방법: 2011년 1월부터 2012년 12월까지 경북대학교병원에서 subarachnoid hemorrhage로 내원한 환자를 대상으로 Angio, CT angio 영상을 기준으로 aneurysm의 개수가 2개 이상인 환자를 대상으로 aneurysm의 rupture 유무, 크기, 모양에 대해서 알아보았다. Aneurysm의 모양은 각각의 특성에 따라 S a , S w , L a , L d , F a 로 나누어 분류하였다. Aneurysm의 rupture 유무는 Neuro intervention 영상의학전문의에 자문으로 이루어졌다. 결과: 2011년 1월부터 2012년 12월까지 SAH로 내원한 환자 모두는 281명이었으며 그 중 두 개이상의 aneurysm이 발견된 환자는 모두 58명이었다. 이 58명의 환자 중 aneurysm이 rupture된 부위는 A-com에서 16명 (27.4%), Rt.P-com에서 10명 (17.1%), Lt.P-com 7명 (11.9%), Rt.MCA bifurcation부위에서 10명 (17.1%), Lt.MCA bifurcation부위에서 5명 (8.5%)으로 나타났다. Rupture aneurysm 평균 size는 6.71mm이었으며, rupture aneurysm의 모양은 제각각이었으나, La, Ld 분류에 포함되는 aneurysm이 대부분이었다. 결론: 두개 이상의 rupture aneurysm이 발견된 환자들 중 가장 많은 rupture가 발병된 부위는 A-com, Rt.P-com, Rt.MCA bifurcation, Lt.P-com, Lt.MCA bifurcation 순이었으며, rupture된 모양은 대부분 lobulating이거나 daughter sac을 가지고 있었음을 알 수 있었다.
Cite
Citations (0)
Routines for Hough transform for line detection use multiple parameters that can be difficult to determine sometimes. Even when a good deal of effort is spent on determining these pa-rameters, the true positives found are often accompanied by some false positives. In this paper, we propose a method called constructive testing for post-processing the lines returned by Hough transform routines with the objective to eliminate false positives. Given a detected line, constructive testing builds a small set of parallel lines based on it. Then, the detected line's distinctiveness as a line in contrast to the other constructed par-allel lines is computed with sample statistics. The detected lines are accepted or rejected based on their distinctiveness. Experimental results show determining a threshold of distinc-tiveness is intuitive and easy and that it effectively eliminates a large number of false positives.
Optimal distinctiveness theory
Hough Transform
Constructive
Line (geometry)
True positive rate
Cite
Citations (0)
A new satellite image classification method using a classifier integration model(CIM)is proposed in this paper. CIM does not use the entire feature vectors extracted from the original data in a concatenated form to classify each datum, but rather uses groups of features related to each feature vector separately. In the training stage, a confusion table calculated from each local classifier that uses a specific feature vector group is drawn throughout the accuracy of each local classifier and then, in the testing stage, the final classification result is obtained by applying weights corresponding to the confidence level of each local classifier. The CIM is applied to the problem of satellite image classification on a set of image data. The results demonstrate that the CIM scheme can enhance the classification accuracy of individual classifiers that use specific feature vector group.
Contextual image classification
Feature vector
Linear classifier
Margin classifier
Confusion
Cite
Citations (5)
Objective To discuss the DSA imaging characteristic and classification of the false aneurysm and true aneurysm complex at the rupture position after intracranial aneurysmal subarachnoid hemorrhage. Methods CT scan and DSA imaging were performed in 50 patients (50 ruptured aneurysm). Results 50 ruptured aneurysm showed SAH by different amount of hemorrhage. DSA test all showed the forming of false aneurysm at the rupture position, tiny, round or near round irregular false aneurysm cavity can be found outside the true aneurysm cavity. According to the DSA imaging presentation we can divide them into three types: 11 cases were type A (near teapot), 27 cases were type B (near gourd), 12 cases were type C (near dumbbell). Conclusion The false aneurysm can form around the rupture position after intracranial rupture, the false aneurysm cavity and true aneurysm cavity together form a special aneurysm complex, it's imaging characteristic under DSA can be divided into three types.
Cite
Citations (0)
In this paper, it will be explained deeply in how to test the measurement of Bedighasan application which is used non linear static analysis medhod against Findbugs which is used static analysis too but it adapts linear code scan. Findbugs is able to detects many bugs pattern, but some parts are not quite accurately detected. It is called false positives when there is no bugs but the system raised warning. It is called false negatives if there is a bugs but the system didn’t raised warning. Here in this paper, with another method, will try to improve false negatives and also false positives. In this paper those 2 variabel will used as measurements. JDBC used as an area of testing because most of JDBC syntax will not seen error unless it executed. This method is needed to handle bugs pattern which is not able to be detected by standard compiler. In this paper will be shown about how bedhigasan will win the false negatives of findbugs and how bedhigasan will loose the false positives of findbugs. Because of the increasing level of awareness of bugs pattern, it decrease the false positives. This application is still need improvement in how mutting some detected bugs.
True positive rate
False positive rate
Cite
Citations (0)
Introduction
About 85% of the non-traumatic SAH are caused by ruptured aneurysms1. Identifying those aneurysms as the bleeding cause is essential for further therapy.Aim of study
The study evaluates the detection of cerebral aneurysm in unenhanced CT images of patients with subarachnoid hemorrhage by a relative hypodense structure in the hyperdense bleeding, the sparing aneurysm sign (SAS).Methods
Three neuroradiologic experienced radiologists rated the aneurysm location and size by applying the SAS in 50 CT-examinations of patients with aneurysmal SAH who underwent an initial CT scan followed by a DSA. The results were analyzed for correlations between aneurysm location, aneurysm size, Fisher-score and the detectability of a SAS. Further a quantitative analysis of the average HU of the aneurysm and the SAH was performed.Results
In 75% of the cases the aneurysm was identified correctly just using the SAS, influenced significant by aneurysm location (p=0.019), Fisher-Score (p=0.008) and aneurysm size (p=0.017). The highest rate of aneurysm detection was given for MCA- (90%) and BA-aneurysms (90%), followed by ACOM- (80%) and ACA-aneurysm (80%). The measured aneurysm size in the correctly identified aneurysm significantly corelates to the size measured in the DSA (p<0.001) and a cut point of 51 HU discriminates aneurysm from SAH with a specificity of 92% and a sensitivity of 86% (Youden´s index 0.78).Conclusions
The aneurysm location and size can be determined in unenhanced CT images in many cases using the new SAS (sparing aneurysm sign). Whereas quantitative measurements of HU can support the aneurysm detection.References
Connolly ES, Rabinstein AA, Carhuapoma JR, et al. Guidelines for the Management of Aneurysmal Subarachnoid Hemorrhage. Stroke. 2012;43. Do you have any conflict of interest to declare?: NoCite
Citations (1)
True positive rate
False Negative Reactions
False Discovery Rate
False positive rate
Cite
Citations (47)
Several studies have evaluated false positives and false negatives produced with partial interval recording (PIR) and momentary time sampling (MTS) using simulated data. However, no study to date has evaluated false positives and negatives using a large sample of non‐simulated behaviors. In addition, few studies have evaluated whether interval methods of data collection alter trends that are evident in continuous records. We conducted three experiments to evaluate the extent to which various interval sizes of MTS and PIR produced false negatives (Experiment 1), false positives (Experiment 2), and trends that were inconsistent with the continuous records (Experiment 3). Collectively, the results show the following: (i) 10‐s PIR and 10‐s MTS produced few false negatives and few false positives (i.e., both were sensitive) to changes in duration events; (ii) 10‐s PIR produced very few false negatives, but an unexpected high percentage of false positives for frequency events; and (iii) each interval size of PIR and MTS produced a high percentage of changes in trending for duration events and frequency events. We briefly discuss the potential limitations and clinical implications of these findings. Copyright © 2012 John Wiley & Sons, Ltd.
True positive rate
Negative
Sampling interval
False Negative Reactions
Sample (material)
Cite
Citations (12)
Evidential Reasoning Based Classifier Combination for an Optimal Remote Sensing Image Classification
The work presented here addresses the problem of the enhancement of the remote sensing image classification. For that, an evidential reasoning based classifier combination method is proposed. The originality of the work lies in the fact that the method treats the outputs of the classifiers by completely ignoring the internal characteristics of the latter. The method is thus general and applicable to any type of classifier. It cumulates the advantages of each classifier without cumulating the disadvantages of them. It overcomes the disadvantages of the remote sensing image classification methods developed in the literature. Thus, it constitutes a powerful tool for several remote sensing image processing applications
Contextual image classification
Cite
Citations (0)
We report a case of unruptured “true posterior communicating artery” (Pcom) aneurysm associated with ruptured internal carotid-posterior communicating artery (IC-PC) aneurysm in a 72-year-old woman, treated successfully by direct surgery. Left internal carotid angiography and three-dimensional CT angiography revealed an IC-PC aneurysm 8 mm in maximum diameter protruding inferolaterally, with a small bulge on the Pcom side which was regarded as part of the IC-PC aneurysm. Intraoperative inspection revealed that the prominence initially thought to be part of the IC-PC aneurysm was a true Pcom aneurysm forming a kissing aneurysm pair with the IC-PC aneurysm. The 2 aneurysms were completely dissected and both were clipped. This is the first reported case of kissing aneurysms of the IC-PC aneurysm and true Pcom aneurysm. A true Pcom aneurysm forming kissing aneurysms with IC-PC aneurysm, as in the present case, may be difficult to diagnose preoperatively. Careful intraoperative observation and complete dissection are therefore important for a good outcome.
Posterior communicating artery
Cite
Citations (5)